Like any scientific technique, radiocarbon dating has limitations, and its results cannot be interpreted uncritically. The archaeological record of Childers, a Late Woodland site in eastern North America, and inferences concerning its occupational history are evaluated here against radiocarbon dates from the site. The record suggests a single, relatively brief, occupation, but radiocarbon-dating results suggest either a much longer continuous occupation or a long series of shorter ones. The apparent conflict between the archaeological record and radiocarbon results is resolved by considering context and integrity of radiocarbon samples, as well as the probabilistic character of the radiocarbon method itself. Considerable dispersion in dating results can occur even in relatively brief occupations, casting doubt on the uncritical interpretation of raw radiocarbon results. Childers's occupational history and chronological placement have important implications for regional culture process during the early Late Woodland interval, and suggest a time lag in the acceptance of cultural innovations.